DECISION THEORY | May 13th 2008

Brett Ryder

The less you know, the more wisely you seem to choose. Helen Joyce rummages through the mental toolbox you use when the facts are scarce ...

From INTELLIGENT LIFE magazine, Spring 2008

"Set down all the Reasons, pro and con, in opposite Columns," wrote Benjamin Franklin in 1779 to his nephew, who was attempting to choose which of two women to propose to. "When you have considered them two or three days...observe what Reasons or Motives in each Column are equal in weight, one to one, one to two, two to three, or the like, and when you have struck out from both sides all the Equalities, you will see in which column remains the balance."

If you have been faced with a difficult decision--which house to buy; whether to accept a posting abroad--you may have done something similar yourself. And you may have had the following strange experience. You listed, you weighed, you calculated the answer--and, in a flash of insight, you realised it was the wrong one.

Did you omit some small but vital factor from one of your Columns? And how on earth did your subconscious get it right so fast?

The answer to these puzzles, says Gerd Gigerenzer, a German psychologist, lies in the way we make decisions, which is not how Franklin--or modern students of decision theory--think we should. Gigerenzer was one of the researchers whose studies of human cognition underpinned Malcolm Gladwell's 2005 best-seller, "Blink", which was about how snap decisions often seem to yield better results than careful analysis. In his new book, "Gut Feelings", Gigerenzer describes some of the quick-and-dirty decision-making tools our brains come fitted with--an "adaptive toolbox" of tricks that we skilfully, and usually unconsciously, pick for the task at hand.

Do you position yourself to catch a ball by solving differential equations in your head, on the fly? Not at all: studies of cricket and baseball players have shown that they use a rule of thumb. If the ball is rising, fix your gaze on it, start running and adjust your running speed so that the image of the ball rises at a constant rate; if it is already high in the air, adjust your running speed so that the angle of your gaze stays constant. Experienced players use this rule without thinking, and do worse if instead they try to work out where the ball is going to land and run there.

We also have rules of thumb for taking decisions, and Gigerenzer's paradox is that knowing a bit is sometimes better than knowing a lot. He describes an experiment in which American and German students had to pick the larger of pairs of American cities: San Diego or San Antonio? Detroit or Milwaukee? Many Americans got the answers wrong, but almost none of the Germans did. They correctly named the first of each pair, even though many of them had never even heard of Milwaukee or San Antonio. That ignorance was informative: the city they had heard of was probably the bigger one.

In another experiment 100 pedestrians were stopped and asked which of 50 company stocks they recognised. The portfolio this generated beat 88% of all entrants in a stock-picking competition. "Ignorance isn't random; it's systematic," says Gigerenzer. "If you know too much, it is harder to distinguish between what is important, and what is not."

When you do have a great deal of expertise, using it all can be counterproductive. The commentators who explain what caused share-price movements use their knowledge of recent stockmarket history to come up with good stories. But when they try to predict prices, many past influences have ceased to matter. A model that includes them all can be outperformed by one that picks out the few most important elements.

Gigerenzer describes how this approach has been used in A&E units to make better decisions about patients with severe chest pains. If a heart attack is imminent, intensive care is best; if not, it wastes money and exposes patients to dangerous infections. The trouble is that telling which patients are most at risk is not easy. But asking just a few, well chosen, yes-or-no questions in sequence (anomalous ECG? Then intensive care. Otherwise, on to next question) does better than either doctors' intuitions, or complex models including scores of probabilities.

Ideally, when it comes to making a decision, you should know the most relevant facts--and only those. But that's not much help if you have to make a decision when you are an expert. Then you must use your natural information-filters, by taking your first thoughts seriously and not giving yourself too much time to have second ones. Good golfers do better if they give themselves little time to prepare for their shots--and even better if they are distracted by researchers asking them to count irregularly spaced sounds. By contrast, beginners do better if they take time to remind themselves of their teacher's maxims first, and are allowed to concentrate.

If you aren't handicapped by knowing too much, don't waste time gathering reams of information. Ask your experts for two-minute briefings, or the top three points. Instead of asking for advice, quiz them about what they would do if they were in your place. Experts have a nasty tendency to cover their backs by telling you everything, and recommending that you leave no stone unturned. This is what Gigerenzer did himself when he went with his mother to see an ophthalmologist about her blindness. The doctor set out the various options, and sounded quite positive about an experimental treatment. But when asked what he would advise his own mother to do, he said: Nothing. Nothing at all.

Once you have finally taken your decision, you can use it to help you find out what really matters--to weight those "Reasons, pro and con" in their Columns. When you have chosen which house to buy, you may realise that a big garden is essential. And turning down that foreign posting could tell you--perhaps to your rueful surprise--that your children's education matters more to you than your career.

(Helen Joyce is education correspondent of The Economist)